Current Biology Vol 23 No 17R716

digestive fluid will allow optimaldigestion of a wide range of proteinand other substrates.

From this and other work on theVenus flytrap, it is possible to begin tobuild a picture of the coordination ofprocesses leading from fast sensoryperception of touch to digestion andabsorption of nutrients (Figure 1).This study also broadens ourknowledge of nutrient transport, andNH4

+ transport in particular, from bothfunctional and evolutionaryperspectives. A number of importantquestions remain: What happens tothe NH4

+ that is absorbed by the trapcells? Is it processed by these cells ortransported to other cell types forprocessing? What are the similaritiesand differences between this NH4

+

transport system and those from plantroots? How are other nutrients (e.g. P,Fe) dealt with? What other signals areinvolved in the coordination of thesecomplex processes? The continued

application of single-cell andwhole-organ physiology andbiochemistry along with genomesequencing and functional genomicsapproaches will ensure that these andother questions continue to beaddressed.

References1. Darwin, C. (1875). Insectivorous Plants

(New York, NY, USA: D Appleton & Co.).2. Robins, R.J., and Juniper, B.E. (1979). The

secretory cycle of Dionaea muscipula Ellis. I.The fine structure and the effect ofstimulation on the fine structure of thedigestive gland cells. New Phytol. 86,279–296.

3. Volkov, A.G., Tejumade, A., Markin, V.S., andJovanov, E. (2008). Kinetics and mechanism ofDionaea muscipula trap closing. Plant Physiol.146, 694–702.

4. Forterre, Y., Skotheim, J.M., Dumais, J., andMahadevan, L. (2005). How the Venus flytrapsnaps. Nature 433, 421–425.

5. Escelante-Perez, M., Krol, E., stange, A.,Geiger, D., Al-Rasheid, K.A.S., Hause, B.,Neher, E., and Hedrich, R. (2011). A special pairof phytohormones controls excitability, slowclosure and external stomach formation in theVenus flytrap. Proc. Natl. Acad. Sci. USA 108,15492–15497.

6. Wheeler, G.L., and Brownlee, C. (2008).Calcium signalling in plants and green algae:changing channels. Trends Plant Sci. 13,506–514.

7. Scherzer, S., Krol, E., Kreuzer, I., Kruse, J.,Karl, F., von Ruden, M., Escalante-Perez, M.,Muller, T., Rennenberg, H., Al-Rasheid, K.A.S.,et al. (2013). The Dionaea muscipulaammonium channel DmAMT1 provides NH4

+

uptake associated with Venus flytrap’s preydigestion. Curr. Biol. 23, 1649–1657.

8. Schulze, W.X., Sanggaard, K.W., Kreuzer, I.,Knudsen, A.D., Bemm, F., Thøgersen, I.B.,Brautigam, A, Thomsen, L.R., Schliesky, S.,et al. (2012). The protein composition of thedigestive fluid from the Venus flytrap shedslight on prey digestion mechanisms. Mol. CellProteomics 11, 1306–1319.

9. Schultze, W.X., Frommer, W.B., and Ward, J.M.(1999). Transporters for ammonium, aminoacids and peptides are expressed in pitchers ofthe carnivorous plant Nepenthes. Plant J. 17,637–646.

10. Rea, P.A. (1982). Fluid composition and factorsthat elicit secretion by the trap lobes of DionaeamuscipulaEllis. Z.Pflanzenphysiol.108, 255–272.

Director, Marine Biological Association of theUK, Citadel Hill, Plymouth PL1 2PB, UK.E-mail: cbr@MBA.ac.uk

http://dx.doi.org/10.1016/j.cub.2013.07.026

Speed Control: Spinal Interneuronswith Crossed Purposes

A recent study has revealed that different populations of commissural spinalinterneurons ensure limb alternation at different speeds of locomotion.

Evdokia Menelaouand David L. McLean

While every journey begins with asingle step, it is the subsequentalternating ones that make the trippossible. Our current understanding ofhow this is achieved in limbed animalswas first articulated over a century ago[1,2]. In his ‘half-center’ hypothesis,Thomas Graham Brown predicted thatnetworks of neurons in the spinal cordwould be organized antagonistically,like themuscles and limbs they control.A recent study by Talpalar et al. [3] hasnow identified fundamental crossing,or commissural, components of thehindlimb ‘half-centers’ and revealedsurprising differences in theircontribution to left–right alternationdepending on how fast the animal istrying to move.

Mice, likemany tetrapods,move overa range of speeds using alternatinggaits. Left–right hindlimb (and forelimb)

alternation requires that flexors on oneside of the body are silent as those onthe other side are active. The same istrue for extensors. This pattern isreinforced by mutual antagonismbetween flexors and extensors on thesame side of the body. To examine thecircuit basis for left–right alternation,Talpalar et al. [3] focused on a singlegenetically identified population ofcells that are known to havecommissural processes. So-called V0neurons arise from the p0 progenitordomain, and contain both excitatory(glutamate/acetylcholine) andinhibitory (GABA/glycine)subpopulations [4,5]. While all V0neurons are defined by the expressionof the Dbx1 transcription factor inprogenitor cells, the V0 population canbe subdivided into Pax7-derived dorsal(V0D) inhibitory and Pax7-negativeventral (V0v) excitatory subgroups(Figure 1A). Talpalar et al. [3] tookadvantage of these differences in

transcription factor expression andtransmitter phenotype and, through aclever use of intersectional geneticapproaches, were able to selectivelyeliminate subsets of V0 interneuronsand examine the effects on locomotorbehavior.As a first pass, Talpalar et al. [3]

eliminated the entire V0 population. Todo so, they selectively killed off the V0cells by expressing a toxin that wasdriven by Dbx1. To confirm that theapproach was working, they used anarray of genetic markers to identify V0cells and found a substantial reductionin their number, while those derivedfrom other progenitor domains werespared. Once they established thespecificity of the ablation approach,next on the list was an examination ofthe consequences. Remarkably,V0-ablated mice survived theprocedure, which provided a uniqueopportunity to test the effect in freelybehaving animals. As you mightexpect from previous work [6],V0-ablated mice lacked the abilityto generate normal alternatinglimb movements, and insteadhopped very much like a rabbit(Figure 1B). Critically, this type ofbehavior is never observed inwild-type mice.

A

B Left–right alternation

Hopping gait

V0

pd1

pd6

p1

p2

p3

pMN

Dorsal

Ventral

Left Right

p0p0p0p0p0p00p0p0p0

Slow Medium Fast

Current BiologyLocomotor speed

C

V0D

V0V

V0D

V0V

V0

Locomotor pattern

V0

Dbx1+

V0

V0D

V0V

Evx1+

Pax7+

Pax7-

Inhibitory

Excitatory

Evx1-

Figure 1. Involvement of V0-derived excita-tion and inhibition in left–right alternation.

(A) The spinal cord is divided into dorsal(pd1–pd6) and ventral (p0–p3; interneurons,pMN, motoneurons) progenitor domains. Thenew study [3] focused on the V0 populationderived from p0 progenitor cells marked byexpression of the transcription factor Dbx1.Dorsal Pax7-derived V0D cells are inhibitoryand ventral V0V cells are excitatory and ex-press the transcription factor Evx1. The V0Vsubdivisionalso includes thePitx2+V0c subsetof cells. (B)Schematicof the locomotorpatternin freely walking mice demonstrates left–rightalternation of opposite limbs in intact mice.V0-deficient mice do not alternate their limbsand hop instead. For simplicity, only thehindlimb pattern is highlighted. The dashedgrey arrows indicate the alternating (top) andsynchronous (bottom) forward progression ofthe hindlimbs. (C) Schematic representationsof the locomotor patterns reported from intactand mutant mice. To the left, the ablatedV0 population is marked by a dashed crossand to the right, diagonal and verticalball-and-stick lines indicate alternation andsynchronous activation of the limbs, respec-tively. Intact mice alternate their limbs andV0-ablated mice are hopping at all locomotorspeeds. Deletion of inhibitory V0D cellsimpairs left–rightalternationat slow tomediumspeeds but not at fast speeds. Ablation ofexcitatory V0V cells leads to the oppositephenotype.

DispatchR717

Because V0 cells were ablatedthroughout the entire nervous system, itwas not clear whether the hoppingphenotype was a result of disruption ofspinal locomotor networks alone. Toaddress this, Talpalar et al. [3] isolatedthe mouse lumbar spinal cord andrecorded the activity of motor nervesduring locomotor-like activity evokedbyneuroactive chemicals. The outcomewasexactlywhatonewouldexpect fromthe mouse hopping behavior; hindlimbmotor nerves burst synchronouslyinstead of in an alternating fashion inV0-ablated mice. Although this wasgoodevidenceby itself, theauthors tookthe extra precaution of geneticallyrestricting V0 ablation to the caudalreachesofspinalcord.Again, consistentwith the importance of V0 spinalinterneurons in maintaining alternation,this more localized perturbationgenerated hopping in the hindlimbs, butnot the forelimbs.

Having confirmed the importance ofspinal V0 cells in left–right alternation,the next step was to investigate therelative contribution of crossedinhibition (V0D) versus crossedexcitation (V0V). Talpalar et al. [3] beganby ablating the V0D inhibitorysubpopulation. Previous work stronglyimplicated crossed inhibition in thecontrol of left–right alternation [6].Consistent with these findings, duringchemically-evoked locomotor-likeactivity in the isolated spinal cord,alternation was absent in mice lackingthe inhibitory V0D subpopulation.However, something that had beencompletely overlooked until now wasthat this effect was specific to slowlocomotion. As the locomotor rhythmincreased in frequency, normalalternation emerged (Figure 1C).Incredibly, when the authors eliminatedthe V0V excitatory population, theyfound the opposite pattern; left–rightalternation was clear at slow speedsbut was lost at fast speeds (Figure 1C).

The story that takes shape is assurprising as the data are convincing.V0 inhibitory commissural interneuronsare critical at slow speeds oflocomotion, but are dispensable atfaster speeds. As the mice move morequickly, an excitatory commissuralpopulation of V0 interneurons takesover responsibility for maintainingleft–right alternation. Given theirsubstantial body of work investigatinghindlimb locomotor circuits [7,8],Talpalar et al. [3] propose a wiringdiagram that likely explains the results

(Figure 2). In this scheme, V0D neuronsdirectly inhibit motoneurons on theopposite side of the body, while V0Vneurons inhibit motoneurons viaactivation of a local intermediary.Therefore, left–right alternation cannowbe explained by two discrete functionalmodules: one that is active at slowspeeds and utilizes crossed inhibition,and one that is engaged at fast speedsand utilizes crossed excitation. Thisobservation challenges traditionalviews regarding left–right alternationand the presumed importance of purelycommissural inhibition [2].

One of a number of open questions ishow these two crossed pathways areengaged at different speeds. Inzebrafish, where different subsets ofinterneurons are also engaged atdifferent speeds, the neurons that areactive at slow speeds are inhibited asthe faster ones are recruited [9].Something similar could be happeningin mice, where V0D cells are activelyinhibited at fast speeds, so as not tointerfere with alternation driven by theV0V cells. Alternatively, each modulecould be excited by distinct pathways.There is reasonable evidence that V2aneurons, derived from the p2 domain,make ipsilateral excitatory connectionspreferentially to the V0V cells [10].Interestingly, genetic ablation ofV2a neurons disrupts alternationspecifically at faster speeds [10,11].This observation is consistent withV2a cells providing a selective sourceof drive to V0V cells (Figure 2). It is stillnot clear what the source of excitatorydrivemay be to the slower V0D cells, butthere are numerous potentialcandidates [4,5].

Now, you may be asking yourself,does this mean rabbits and kangarooslack V0 neurons? It is, of course,a possibility that during evolutionthere was a complete loss or partialcompromise of V0 connections.Another possibility is that the crossedconnections are subject toneuromodulation, which alters theirinfluence on contralateral motor pools.In rabbits, locomotor output fromreduced or isolated spinal cordpreparations is always consistent withhopping movements [12]. However, innewborn rats, the connections fromcommissural neurons known as ‘switchcells’ to motoneurons can beconverted from polysynaptic inhibitoryto monosynaptic excitatory in thepresence of serotonin [13]. Dialing upor down the strength of crossing

Left Right

Current Biology

Excitation

Inhibition

V0D

V0V

V2a

?

IIN

MN

EINCIN

Figure 2. Proposed speed-related organization of networks involved in left–right alternation.

Wiring diagram highlights the two neuronal modules involved in alternation at different speeds.At slow speeds (red lines), the V2a interneurons drive the excitatory V0V subpopulation toinhibit contralateral MNs via the activation of IINs. At fast speeds (blue lines), the inhibitoryV0D cells are activated by ipsilateral interneurons of unknown identity (?) to directly inhibitcontralateral MNs. EIN, excitatory interneuron; CIN, commissural interneuron; IIN, ipsilateralinterneuron; MN, motoneuron. For more details see main text.

Current Biology Vol 23 No 17R718

connections via neuromodulatorswould certainly provide a more flexibleand less permanent means to generatedifferent gaits.

Given the phylogenetic conservationof the transcription factor code forspinal differentiation [4,5] and theobservation that interneuron switchingoccurs not only in mice as describedhere but also in fish [9], it is more thanlikely that similar mechanisms are atplay within our own spinal cord. Assuch, the work by Talpalar et al. [3]brings us several steps closer to the

resolution of a journey that began along time ago. Or, at least in the caseof V0-deficient animals, several hopscloser.

References1. Graham Brown, T. (1911). The intrinsic factors

in the act of progression in mammals. Proc. R.Soc. Lond. B 84, 308–319.

2. Grillner, S. (2006). Biological pattern generation:the cellular and computational logic of networksin motion. Neuron 52, 751–766.

3. Talpalar, A.E., Bouvier, J., Borgius, L.,Fortin, G., Pierani, A., and Kiehn, O. (2013).Dual-mode operation of neuronal networksinvolved in left-right alternation. Nature 500,85–88.

4. Arber, S. (2012). Motor circuits in action:specification, connectivity, and function.Neuron 74, 975–989.

5. Goulding, M. (2009). Circuits controllingvertebrate locomotion: moving in a newdirection. Nat. Rev. Neurosci. 10, 507–518.

6. Lanuza, G.M., Gosgnach, S., Pierani, A.,Jessell, T.M., and Goulding, M. (2004). Geneticidentification of spinal interneurons thatcoordinate left-right locomotor activitynecessary for walking movements. Neuron 42,375–386.

7. Dougherty, K.J., and Kiehn, O. (2010).Functional organization of V2a-relatedlocomotor circuits in the rodent spinal cord.Ann. NY Acad. Sci. 1198, 85–93.

8. Kiehn, O. (2006). Locomotor circuits in themammalian spinal cord. Annu. Rev. Neurosci.29, 279–306.

9. McLean, D.L., Masino, M.A., Koh, I.Y.,Lindquist, W.B., and Fetcho, J.R. (2008).Continuous shifts in the active set of spinalinterneurons during changes in locomotorspeed. Nat. Neurosci. 11, 1419–1429.

10. Crone, S.A., Quinlan, K.A., Zagoraiou, L.,Droho, S., Restrepo, C.E., Lundfald, L.,Endo, T., Setlak, J., Jessell, T.M., Kiehn, O.,et al. (2008). Genetic ablation of V2a ipsilateralinterneurons disrupts left-right locomotorcoordination in mammalian spinal cord. Neuron60, 70–83.

11. Crone, S.A., Zhong, G., Harris-Warrick, R., andSharma, K. (2009). In mice lacking v2ainterneurons, gait depends on speed oflocomotion. J. Neurosci. 29, 7098–7109.

12. Vidal, C., Viala, D., and Buser, P. (1979).Central locomotor programming in the rabbit.Brain Res. 168, 57–73.

13. Butt, S.J., and Kiehn, O. (2003). Functionalidentification of interneurons responsible forleft-right coordination of hindlimbs inmammals. Neuron 38, 953–963.

Department of Neurobiology, NorthwesternUniversity, Evanston, IL 60208, USA.E-mail: david-mclean@northwestern.edu

http://dx.doi.org/10.1016/j.cub.2013.07.064

Active Vision: Adapting How to Look

A new study has found that artificial occlusion of central vision leads to rapidemergence, and long-term maintenance of a new preferred retinal locus offixation. These findings have important implications for the understanding ofvisual and oculomotor plasticity as well as for the development of rehabilitationtechniques.

Martina Poletti1

and Michele Rucci1,2

Finding a needle in a haystack is anotoriously difficult task. Part of thedifficulty originates from thenon-uniform resolution of the visualsystem. Even though the human eyecovers a broad field, only a regionsmaller than one degree in visualangle — approximately the size of athumb at arm’s distance — offers theresolution necessary for seeing finedetail and distinguishing needles fromhay. This is the portion of the scene

that projects onto the central fovea, adepression in the retinal surface wherereceptors are most densely packed.Not surprisingly, humans normally usethis region as their preferred retinallocus for acquiring fine spatialinformation and move this locus fromone point of interest to the next bymeans of very fast eye movements(saccades). But what happens whenthis preferred retinal region suddenlybecomes unusable? A new study byKwon et al. [1], reported in this issueof Current Biology, shows that normal,healthy observers rapidly adapt to an

artificial obstruction of the fovea bydeveloping a new preferred retinallocus, which they then retain evenafter relatively long periods of normalunobstructed vision.Imagine being at The Louvre looking

at La Gioconda (Figure 1). At a distanceof approximately one meter from thepainting, only an area of a few squaredcentimeters falls within the fovealregion with the highest visualresolution. As an observer with normalvision (observer A) looks atMona Lisa’sleft eye, the rest of the painting appearsblurred, the degree of blurringincreasing with the distance from thecurrent point of fixation. To examineMona Lisa’s mouth (is she reallysmiling?), the observer will need tomove his eyes so to bring the regionof interest on the fovea. The mouthwill then become visible at the highestlevel of detail and — perhaps, for this